Cervical cancer is one of the most common malignancies in women worldwide, second only to breast cancer in both incidence and mortality. In the United States, both morbidity and mortality have decreased dramatically since the 1940s, when the Papanicoleau (Pap) smear test was introduced (xe2x80x9cNIH Consensus Statement (1) Cervical Cancerxe2x80x9d, infra). Despite aggressive screening there are still approximately 16,000 new cases of cervical cancer per year, and approximately 5,000 deaths. Although part of the reason for this level of morbidity and mortality is due to the fact that some women fail to have regular routine Pap smears; another reason is that Pap smears are relatively frequently misdiagnosed. This includes inherently high false negative rates and inconsistent cytological reporting systems (Koss, L. G., 1989, infra); Yobs, A. R., et al, infra; and Koss, L. G., 1993, infra).
A major problem has been that since the introduction of Pap smear screening, the descriptive terminology for cervical cytology has been variable and confusing (Jones, H. W., infra). In 1988, a NCI-sponsored workshop developed new terminology for Pap smear screening that was intended to clarify the system of cytologic scoring, which resulted in xe2x80x9cThe 1988 Bethesda System for reporting cervical/vaginal cytological diagnosesxe2x80x9d (National Cancer Institute Workshop, infra). This was further modified in 1991 (Broder, S., infra). The Bethesda System (TBS), as it became known, introduced the categories of xe2x80x9catypical squamous cells of undetermined significancexe2x80x9d (ASCUS), and xe2x80x9catypical glandular cells of undetermined significancexe2x80x9d (AGCUS or AGUS). In our previous patent application Ser. No. 09/461,930, we provided an effective method for dealing with the AGUS diagnostic category. However, the ASCUS xe2x80x9cdiagnosticxe2x80x9d category remains problematic for the cytopathologist and practicing physician.
The original intent was to use the ASCUS designation as a category of squamous cell abnormality for cases not diagnostic of a reactive/inflammatory, dysplastic, or neoplastic condition (Boemer, S. L., et al, infra). Unfortunately, the ASCUS diagnosis does not identify a specific clinical entity, and reflects one of the inherent limitations of the Pap screening procedure. To add to this problem, the diagnosis of ASCUS rapidly grew popular in testing laboratories and has approached frequencies up to 10% of all Pap test results! The ASCUS diagnosis has been termed xe2x80x9ca wastebasket category in some labsxe2x80x9d and xe2x80x9can I-don""t-know categoryxe2x80x9d (xe2x80x9cGetting a handle on ASCUS; a new clinical trial could show howxe2x80x9d, JNCI 87(No. 7), Jun. 7, 1995. Currently, it is recommended that the diagnosis of ASCUS should be qualified additionally, to include ASCUS-favor reactive, ASCUS-not otherwise specified (NOS), or ASCUS-favor squamous intraepithelial lesion (SIL) (Kurman, R. J., Hanson, et al, infra). However, these distinctions have been controversial; the uncertainty associated with a diagnosis of ASCUS has been well documented (Boemer, S. L., et al, infra).
An ASCUS diagnosis presents a serious clinical challenge. Up to one third of women who receive a diagnosis of ASCUS harbor significant lesions. Moreover, more than one third of HSILs found in screening patients are identified from ASCUS Pap test results. However, this also means that two thirds of women who receive a Pap smear diagnosis of ASCUS do not harbor a cervical lesion.
No satisfactory adjunct test has previously been identified that serves to complement the diagnosis of ASCUS, thereby aiding in identifying the presence of a significant lesion. However, screening for the presence of high-risk types of human papillomavirus (HPV) in Pap smear specimens has helped to identify those women with underlying HSIL; see Sherman, M. E., et al, infra; and Hatch, K. D., et al, infra. Molecular and epidemiologic evidence has accrued that indicates that infection with high-risk types of HPV is a major risk factor for invasive cervical cancer; see zur Hausen, H., infra. The association, however, is not absolute because a small proportion of cytologically negative women have detectable high-risk HPV DNA. Conversely, some HSILs are HPV negative; see Boemer, S. L., et al, infra. A large multicenter clinical trial (the ALTS study), designed to determine the utility of HPV DNA testing, is currently in progress, sponsored by the National Cancer Institute.
Recent studies have indicated that expression of proliferation-associated proteins, e.g. Ki67 and PCNA, may also serve as useful indicators of the presence of dysplastic lesions; see Bulten, J., et al, infra; and Mittal, K. R., et al, infra. However, no specific analyses have been applied to Pap smear specimens.
The essence of the problem is that ASCUS diagnoses will include a significant fraction of cases of low grade intraepithelial neoplasia (CINI), also termed low grade squamous intraepithelial lesion (LSIL). Only a small percentage of LSILs will progress to significant lesions, i.e. high grade squamous intraepithelial lesion (HSIL, also designated CIN II, III) and carcinoma in situ (CIS). Another serious problem is that the incidence of high grade lesions may be as high as 10% in patients who receive an ASCUS diagnosis of their cervical/vaginal smearxe2x80x94a total of approximately 2-3 million per year (Davey, D. D., et al, infra). Because of the current legal climate in the United States, the majority of ASCUS diagnoses are aggressively followed up with colposcopy and selected biopsy. Although colposcopy for all patients with atypical smears identifies those patients with SIL, it carries substantial personnel and financial costs, and subjects some patients to unnecessary invasive procedures that carry an element of risk.
It would be highly desirable to provide a reliable supplementary test that will allow for further discrimination of ASCUS diagnoses, identifying those patients with significant lesions, and providing a better prognostic picture of which LSILs are destined to progress versus regress.
One approach is to use a surrogate biomarker. We previously identified such a candidate biomarkerxe2x80x94the MN/CA9 antigen (Liao, S. Y., Brewer, C., et al, infra); and Liao, S. Y., and Stanbridge, E. J., et al, infra). Several years ago, we found that this antigen is expressed on the plasma membrane surface of virtually all cervical carcinomas, irrespective of their provenance (Liao, S. Y., Brewer, C., et al, et al, infra). The gene encoding MN/CA9 was cloned and, at the time, was novel (Pastorek, J., et al, infra). It has now been recognized as a member of the carbonic anhydrase family. Recent studies have suggested that, in addition to carbonic anhydrase activity, MN/CA9 may function as an adhesion protein and preliminary evidence suggests it may have transforming activity in mouse NIH3T3 assays (Zavada, J., et al, infra).
In a study of over 300 Pap smears, representing the complete spectrum of benign and neoplastic cervical lesions, we showed that high levels of MN/CA9 were expressed in all cases of AIS and adenocarcinoma, in more than 90% of cases of cervical squamous cell carcinoma and, to a lesser degree, in SILs. This study concluded the following: 1) MN/CA9 expression in exfoliative cells recapitulates MN/CA9 expression in the tissue sections; 2) diffuse strong MN/CA9 immunoreactivity of cells in the cervical smears always appears to be associated with dysplasia, regardless of whether the positive cells have the cytologic appearance of normal or atypical endocervical cells or dysplastic cells; and 3) virtually all dysplastic glandular cells in the cytologic smears express MN/CA9 protein. Thus, MN/CA9 protein expression in exfoliative endocervical cells would appear to be an important diagnostic biomarker of glandular dysplasia, AIS, invasive adenocarcinoma, and/or SIL (Liao, S. Y, and Stanbridge, E. J., etal, infra).
As discussed above, the cytologic diagnoses of ASCUS and AGUS are particularly vexing for gynecologistsxe2x80x94leaving them in a quandary as to the optimal regimen of further examination/treatment. In our prior, co pending patent application, we demonstrated that the combination of cytology and MN/CA9 expression in AGUSxe2x80x94diagnosed Pap smears allows for the identification of those patients harboring significant lesions (CIN II, III and/or AIS/CA). The results described in the application are so promising that a clinical trial has been initiated under the auspices of the National Cancer Institute and the Multicenter Gynecologic Oncology Group.
The present invention provides an adjunct test to the conventional Papanicolaou (Pap) smear, that will allow for identification of those patients with a diagnosis of ASCUS who harbor a significant lesion (high grade squamous intraepithelial lesion (HSIL) and/or carcinoma). The invention proceeds by immunostaining Pap smears and corresponding histologic biopsies for expression of the MN/CA9 antigen. More specifically, Pap smears with a diagnosis of ASCUS are immunostained for expression of the MN/CA9 antigen. The presence of MN/CA9 immunostaining of the Pap smears is correlated with the histologic diagnosis of the corresponding biopsy. Those ASCUS-diagnosed Pap smears with specific MN/CA9 immunostaining correspond to biopsies that contain SIL and/or carcinoma (CA). The pattern of immunostaining (focal or diffuse staining of atypical cells versus normal cells only) discriminates between biopsies that contain low grade squamous intraepithelial lesion (LSIL) (immunostaining of normal cells only) versus a high grade squamous intraepithelial lesion (HSIL) and/or CA (focal or diffuse staining of atypical cells).